]>
https://gitweb.dealii.org/ - dealii.git/log
Martin Kronbichler [Fri, 10 Jul 2015 16:59:15 +0000 (18:59 +0200)]
Make compile gcc-4.4
Timo Heister [Fri, 10 Jul 2015 16:07:54 +0000 (12:07 -0400)]
Merge pull request #1105 from drwells/fix-step-32-references
Fix some minor doxygen issues in step-32.
Wolfgang Bangerth [Fri, 10 Jul 2015 16:07:05 +0000 (10:07 -0600)]
Merge pull request #1106 from tjhei/fix_tests2
fix tests
Timo Heister [Fri, 10 Jul 2015 16:06:07 +0000 (12:06 -0400)]
fix tests
- 2 tests require functionparser
- update another libc++ typename string replacement
David Wells [Fri, 10 Jul 2015 14:19:19 +0000 (10:19 -0400)]
Fix some minor doxygen issues in step-32.
Doxygen did not pick up the quoted text next to a @ref correctly, so I
reworded the paragraph to not need the quoted text.
Wolfgang Bangerth [Fri, 10 Jul 2015 11:39:58 +0000 (05:39 -0600)]
Merge pull request #1104 from kronbichler/master
Avoid compile error on gcc-4.6 (and probably older).
Martin Kronbichler [Fri, 10 Jul 2015 11:29:42 +0000 (13:29 +0200)]
Avoid compile error on gcc-4.6 (and probably older).
Bruno Turcksin [Fri, 10 Jul 2015 02:17:39 +0000 (21:17 -0500)]
Merge pull request #1103 from drwells/remove-wrapcomments-string-import
Remove a deprecated function usage.
David Wells [Fri, 10 Jul 2015 00:53:26 +0000 (20:53 -0400)]
Remove a deprecated function usage.
This function (string.find) was deprecated in python 2.4 in favor of the
.find() method of strings and was and removed in python 3.
Wolfgang Bangerth [Thu, 9 Jul 2015 20:20:37 +0000 (14:20 -0600)]
Merge pull request #1099 from tjhei/formatting
Formatting
Timo Heister [Thu, 9 Jul 2015 20:07:22 +0000 (16:07 -0400)]
run wrapcomments.py
Conflicts:
include/deal.II/numerics/error_estimator.h
Timo Heister [Thu, 9 Jul 2015 18:29:59 +0000 (14:29 -0400)]
fix to wrapcomments.py
Timo Heister [Thu, 9 Jul 2015 16:55:40 +0000 (12:55 -0400)]
fix doxygen errors
Timo Heister [Thu, 9 Jul 2015 16:31:36 +0000 (12:31 -0400)]
manual formatting
This makes doxygen and wrapcomments.py happy
Wolfgang Bangerth [Thu, 9 Jul 2015 19:58:30 +0000 (13:58 -0600)]
Merge pull request #1097 from tjhei/grid_distort_random_boost
switch GridTools::distort_random to boost::random
Bruno Turcksin [Thu, 9 Jul 2015 19:41:11 +0000 (14:41 -0500)]
Merge pull request #1100 from tjhei/muelu_in_doxygen
make muelu preconditioner appear in doxygen
Timo Heister [Thu, 9 Jul 2015 19:38:10 +0000 (15:38 -0400)]
make muelu preconditioner appear in doxygen
Timo Heister [Thu, 9 Jul 2015 19:25:49 +0000 (15:25 -0400)]
Merge pull request #1083 from tjhei/cell_id_to_str
add CellId::to_string()
Timo Heister [Thu, 9 Jul 2015 18:52:28 +0000 (14:52 -0400)]
move implementation to .cc
Timo Heister [Sun, 5 Jul 2015 17:33:34 +0000 (13:33 -0400)]
clean up tests and use to_string()
Timo Heister [Sat, 4 Jul 2015 11:23:24 +0000 (07:23 -0400)]
add CellID::to_string()
Timo Heister [Thu, 9 Jul 2015 15:54:20 +0000 (11:54 -0400)]
switch GridTools::distort_random to boost::random
Use boost::random instead of std::rand(). This makes calls to the
function deterministic (distortion will always be the same for the same
mesh) and hopefully makes the testsuite more robust (differences in
rand() on MAC OS).
Wolfgang Bangerth [Thu, 9 Jul 2015 14:55:43 +0000 (08:55 -0600)]
Merge pull request #1096 from tjhei/fix_tests
MAC: maybe fix test
Timo Heister [Thu, 9 Jul 2015 14:31:47 +0000 (10:31 -0400)]
MAC: maybe fix test
Forcing all computations to be done as long double with the hope that
this fixes the numerical differences on the MAC tester.
Martin Kronbichler [Thu, 9 Jul 2015 13:46:50 +0000 (15:46 +0200)]
Merge pull request #1072 from kronbichler/master
Introduce vectorized transposition for array-to-structure conversions
Martin Kronbichler [Thu, 2 Jul 2015 15:47:55 +0000 (17:47 +0200)]
Introduce vectorized transposition for array-to-structure conversions
Martin Kronbichler [Thu, 2 Jul 2015 15:47:55 +0000 (17:47 +0200)]
Introduce vectorized transposition for array-to-structure conversions
Martin Kronbichler [Thu, 9 Jul 2015 12:57:52 +0000 (14:57 +0200)]
Merge pull request #1094 from tjhei/fix_tests
fix tests
Timo Heister [Thu, 9 Jul 2015 12:54:52 +0000 (14:54 +0200)]
Merge pull request #1087 from tjhei/clang_warnings
disable clang warnings with cxx11 disabled
Timo Heister [Mon, 6 Jul 2015 08:27:28 +0000 (04:27 -0400)]
disable clang warnings
compiling the bundled boost with clang without cxx11 causes hundreds of
warnings.
Timo Heister [Thu, 9 Jul 2015 12:51:53 +0000 (08:51 -0400)]
fix tests
Wolfgang Bangerth [Tue, 7 Jul 2015 20:20:53 +0000 (14:20 -0600)]
Merge pull request #1091 from drwells/fix-single-digit-output-time
Fix single digit output time
David Wells [Tue, 7 Jul 2015 11:40:41 +0000 (07:40 -0400)]
Remove 'if (true)' statements.
These are redundant and the related comment is no longer correct (there
are, as of the previous commit, no new variables declared in that
scope).
David Wells [Tue, 7 Jul 2015 01:39:16 +0000 (21:39 -0400)]
Fix strange VTK timestamps with refactoring.
VTK files, based on when they are written, may save output with text
like
This file was generated by the deal.II library on 2015/7/6 at 17: 3:49
which, while correct, is a bit strange to read. This behavior may occur
in other output file formats. I fixed this by calling utility functions
with the correct behavior.
Wolfgang Bangerth [Tue, 7 Jul 2015 03:16:04 +0000 (21:16 -0600)]
Merge pull request #1088 from tjhei/remove_warnings
disable muparser warnings
Guido Kanschat [Mon, 6 Jul 2015 14:19:12 +0000 (16:19 +0200)]
Merge pull request #1086 from tjhei/fix_test_output
fix test output
Guido Kanschat [Mon, 6 Jul 2015 14:17:09 +0000 (16:17 +0200)]
Merge pull request #1085 from tjhei/mac_fix_tests
fix tests on MAC OS
Guido Kanschat [Mon, 6 Jul 2015 14:15:39 +0000 (16:15 +0200)]
Merge pull request #1089 from tjhei/clang_warning_pragma_warnings
fix pragma warnings warnings... sigh :-(
Timo Heister [Mon, 6 Jul 2015 09:31:34 +0000 (05:31 -0400)]
fix pragma warnings warnings
PR #1076 disabled warnings due to disabling unknown warnings with a
pragma. This causes the warning
src.cxx:4:3: error: unknown warning group '-Wpragmas', ignored [-Werror,-Wunknown-pragmas]
_Pragma("GCC diagnostic ignored \"-Wpragmas\"")
with Clang 3.4.0. It seems it supports Wunknown-pragmas which is okay to be
used with gcc too. Sigh.
Timo Heister [Mon, 6 Jul 2015 07:41:14 +0000 (03:41 -0400)]
fix tests
- do not use pow(int, int)
- fix unify_pretty_function()
- replace typenames
Timo Heister [Mon, 6 Jul 2015 08:36:11 +0000 (04:36 -0400)]
disable muparser warnings
suppress -Wnested-anon-types warning with clang 3.6 in muParser.
Timo Heister [Mon, 6 Jul 2015 08:07:22 +0000 (04:07 -0400)]
fix test output
This due to the recent changes in the vtk GridOut
Timo Heister [Sun, 5 Jul 2015 17:06:38 +0000 (13:06 -0400)]
Merge pull request #1084 from guidokanschat/fix_2_multigrid_tests
Fix the ordering of renumbering in two tests
Guido Kanschat [Sun, 5 Jul 2015 16:44:40 +0000 (18:44 +0200)]
Fix the ordering of renumbering in two tests
Guido Kanschat [Sun, 5 Jul 2015 15:40:05 +0000 (17:40 +0200)]
Merge pull request #1082 from tjhei/string_replace
new Utilities::replace_in_string() which returns a string with all occurrences of one substring replaced by another
Timo Heister [Sun, 5 Jul 2015 15:16:07 +0000 (11:16 -0400)]
new Utilities::replace_in_string()
Timo Heister [Sun, 5 Jul 2015 15:27:03 +0000 (11:27 -0400)]
Merge pull request #1077 from guidokanschat/improve_svg
Fix font sizes in SVG output
The font scaling in GridOut::write_svg() was broken since the units were missing. It has been fixed and an additional parameter GridOutFlags::Svg::cell_font_scaling has been introduced for tuning.
Guido Kanschat [Sat, 4 Jul 2015 07:19:11 +0000 (09:19 +0200)]
adjust test output
Guido Kanschat [Sat, 4 Jul 2015 06:45:24 +0000 (08:45 +0200)]
Fix font sizes in SVG output
Guido Kanschat [Sun, 5 Jul 2015 15:19:10 +0000 (17:19 +0200)]
Merge pull request #1076 from guidokanschat/diagnostic_pragmas
Remove warnings due to disabling unknown warnings
Guido Kanschat [Sun, 5 Jul 2015 15:06:39 +0000 (17:06 +0200)]
reduce pragma test to the necessary
Timo Heister [Sun, 5 Jul 2015 14:55:38 +0000 (10:55 -0400)]
Merge pull request #1081 from guidokanschat/fix_vtu_tests
Adjust test output for VTU tests to additional manifold id
Guido Kanschat [Sat, 4 Jul 2015 05:22:31 +0000 (07:22 +0200)]
test for all the pramas used in DEAL_II_DISABLE_EXTRA_DIAGNOSTICS to avoid warnings in clang 3.4
Guido Kanschat [Sun, 5 Jul 2015 14:50:00 +0000 (16:50 +0200)]
Adjust test output for VTU tests to additional manifold id
Luca Heltai [Sun, 5 Jul 2015 14:22:26 +0000 (16:22 +0200)]
Merge pull request #1078 from guidokanschat/partial_grid_out
VTK and VTU output with level number, material id, and subdomain ids
Guido Kanschat [Sun, 5 Jul 2015 11:46:06 +0000 (13:46 +0200)]
astyle
Guido Kanschat [Sun, 5 Jul 2015 11:38:17 +0000 (13:38 +0200)]
add manifold id to output and adjust output files
Guido Kanschat [Sun, 5 Jul 2015 11:34:35 +0000 (13:34 +0200)]
remove include
Guido Kanschat [Sun, 5 Jul 2015 10:56:38 +0000 (12:56 +0200)]
ignore mpirun tests without mpi
Guido Kanschat [Sun, 5 Jul 2015 09:42:29 +0000 (11:42 +0200)]
add test for multiple subdomains and convert huge unsigned to negative values
Guido Kanschat [Sat, 4 Jul 2015 12:25:11 +0000 (14:25 +0200)]
VTK and VTU output in GridOut restructured to use iterator ranges internally
Add output of level number, material id, and subdomain ids
Guido Kanschat [Sun, 5 Jul 2015 10:11:40 +0000 (12:11 +0200)]
Merge pull request #1079 from tjhei/testsuite_pretty_function
rework unify_pretty_function() in testsuite so that sed is not needed anymore
Luca Heltai [Sun, 5 Jul 2015 10:08:29 +0000 (12:08 +0200)]
Merge pull request #1080 from tjhei/remove_warning
remove warning
Timo Heister [Sun, 5 Jul 2015 10:00:05 +0000 (06:00 -0400)]
remove warning
gcc gives the nonsensical warning "comparison of unsigned expression < 0
is always false". Work around this.
Timo Heister [Sun, 5 Jul 2015 08:46:10 +0000 (04:46 -0400)]
rework unify_pretty_function() in testsuite
We used to call into sed to do some string replacement in order to
normalize the output of __PRETTY_FUNCTION__. This is crazy and won't
work on MAC OS (sed doesn't accept -e). Do this using string replacement
instead.
Martin Kronbichler [Fri, 3 Jul 2015 15:04:49 +0000 (17:04 +0200)]
Merge pull request #1075 from tamiko/compilation_fixes
Bugfix: Compile fixes for Functions::Polynomial<dim>
Matthias Maier [Fri, 3 Jul 2015 14:55:37 +0000 (16:55 +0200)]
Bugfix: Compile fixes for Functions::Polynomial<dim>
Change dealii::Vector to std::vector to not have to include vector.h in
function_lib.h. This fixes a compilation error encountered with several
compilers.
Further, fix the corresponding test
Luca Heltai [Fri, 3 Jul 2015 13:47:24 +0000 (15:47 +0200)]
Merge pull request #1074 from tamiko/null_operator_cleanup
Restructure null_operator
Matthias Maier [Fri, 3 Jul 2015 13:28:19 +0000 (15:28 +0200)]
Restructure null_operator
This commit restructures the null_operator optimization in LinearOperator
slightly and removes an ambiguous constructor variant.
Martin Kronbichler [Fri, 3 Jul 2015 12:37:37 +0000 (14:37 +0200)]
Merge pull request #1067 from angelrca/Function_Polynomial
New class Function::Polynomial
Martin Kronbichler [Fri, 3 Jul 2015 10:21:02 +0000 (12:21 +0200)]
Merge pull request #1073 from guidokanschat/fix_mac
Fix problems where the library did not compile on MacOS with clang 3.3
Guido Kanschat [Fri, 3 Jul 2015 10:09:45 +0000 (12:09 +0200)]
check for std::isnan and make sure we never use a function that does not exist
Guido Kanschat [Fri, 3 Jul 2015 07:27:24 +0000 (09:27 +0200)]
fix more powers
compiling on Mac with clang 5.0.0 (3.3) again
Guido Kanschat [Thu, 2 Jul 2015 21:03:24 +0000 (23:03 +0200)]
Fix the problems on Mac with isnan and pow
Guido Kanschat [Thu, 2 Jul 2015 20:30:15 +0000 (22:30 +0200)]
Merge pull request #1047 from guidokanschat/new_meshes
Add three new meshes to GridGenerator: simplex, cheese and hyper_cross
Guido Kanschat [Sun, 28 Jun 2015 19:15:46 +0000 (21:15 +0200)]
Add three new meshes to GridGenerator: simplex, cheese and hyper_cross
Luca Heltai [Thu, 2 Jul 2015 14:58:29 +0000 (16:58 +0200)]
Merge pull request #1070 from davydden/step36-eigenvalues
fixed the expression for expected eigenvalues in Step-36
Luca Heltai [Thu, 2 Jul 2015 13:36:15 +0000 (15:36 +0200)]
Merge pull request #1071 from tamiko/restructure_cmake_cxx11
CMake: Restructure logic in check_01_cxx_features
Matthias Maier [Thu, 2 Jul 2015 10:27:28 +0000 (12:27 +0200)]
CMake: Restructure logic in check_01_cxx_features
Denis Davydov [Thu, 2 Jul 2015 08:47:28 +0000 (10:47 +0200)]
fixed the expression for expected eigenvalues in Step-36
angelrca [Wed, 1 Jul 2015 08:44:25 +0000 (10:44 +0200)]
New class Function::Polynomial
New class Function::Polynomial
New class Function::Polynomial
New class Function::Polynomial
Matthias Maier [Thu, 2 Jul 2015 07:40:47 +0000 (09:40 +0200)]
Merge pull request #1065 from drwells/fix-trivially-copyable-cmake-test
[RFC] Fix trivially copyable cmake test
Matthias Maier [Thu, 2 Jul 2015 05:53:21 +0000 (07:53 +0200)]
Merge pull request #1068 from bangerth/add-missing-function
Add a declared but not implemented function.
Wolfgang Bangerth [Wed, 1 Jul 2015 16:20:54 +0000 (11:20 -0500)]
Add a declared but not implemented function.
This addresses a linker error recently posted to the mailing list. The issue
is that we have iterators that represent faces of cells and a specialization
for the faces of 1d cells, i.e., vertices. These class specializations have
most of the regular accessor members declared, but here is one that was
declared without being defined.
This patch adds a definition of the function. The issue is relatively
uncritical since the so-defined function does not actually have to do
anything other than return an invalid object (vertices have no children).
Wolfgang Bangerth [Wed, 1 Jul 2015 04:04:57 +0000 (23:04 -0500)]
Merge pull request #1066 from QiaoLei-88/fix_max_n_cells
Fix max n cells
Lei Qiao [Wed, 1 Jul 2015 03:44:38 +0000 (22:44 -0500)]
update mpi/refine_and_coarsen_fixed_number_07 and it's output.
1. Increase number of cells by global refinement to weaken influence of float truncation error;
2. Because number of cells increased, construct refinement indicator as test _01 to avoid float overflow;
3. Update test drivers interface because it is too complicated to calculate global number of cells;
4. Update reference output.
Lei Qiao [Wed, 1 Jul 2015 01:34:39 +0000 (20:34 -0500)]
fix bug introduce by commit
594dc87 in function GridRefinement::refine_and_coarsen_fixed_number()
David Wells [Wed, 1 Jul 2015 00:15:34 +0000 (20:15 -0400)]
Check for 'std::is_trivially_copyable' correctly.
This should be checked whenever C++11 is enabled; more specifically, it
should be checked when compiling with C++14 too.
David Wells [Wed, 1 Jul 2015 00:15:05 +0000 (20:15 -0400)]
Fix a spelling error in a comment.
Wolfgang Bangerth [Tue, 30 Jun 2015 14:29:36 +0000 (09:29 -0500)]
Merge pull request #1059 from bangerth/fix-isfinite
Verify that a number is not a NaN before testing for finiteness.
Wolfgang Bangerth [Mon, 29 Jun 2015 23:15:23 +0000 (18:15 -0500)]
Add a changelog entry.
Wolfgang Bangerth [Mon, 29 Jun 2015 22:59:54 +0000 (17:59 -0500)]
Verify that a number is not a NaN before testing for finiteness.
While not intuitive, std::isfinite triggers a floating point exception when called
on a signalling NaN. Avoid this.
Matthias Maier [Tue, 30 Jun 2015 14:20:30 +0000 (16:20 +0200)]
Merge pull request #1063 from tjhei/cmake_depend_msg
clarify cmake dependency message
Timo Heister [Tue, 30 Jun 2015 13:36:06 +0000 (09:36 -0400)]
clarify cmake dependency message
change the message:
BLA has unmet configuration requirements: MPI
has to be set to "ON".
to:
BLA has unmet configuration requirements: DEAL_II_WITH_MPI
has to be set to "ON".
Timo Heister [Tue, 30 Jun 2015 13:17:54 +0000 (09:17 -0400)]
Merge pull request #1061 from bangerth/fix-test-6
Avoid computing atan(y/x) where x=0.
Wolfgang Bangerth [Tue, 30 Jun 2015 11:33:59 +0000 (06:33 -0500)]
Merge pull request #1057 from angelrca/Monomial_gradient
Functions::Monomial
Timo Heister [Tue, 30 Jun 2015 10:15:20 +0000 (06:15 -0400)]
Merge pull request #1062 from bangerth/fix-gmres-breakdown
Handle the case of lucky breakdown in GMRES better, without testing whether 1/a is actually finite
Timo Heister [Tue, 30 Jun 2015 10:10:47 +0000 (06:10 -0400)]
Merge pull request #1060 from bangerth/fix-test-5
Avoid setting boundary objects that don't fit.
angelrca [Mon, 29 Jun 2015 15:48:37 +0000 (17:48 +0200)]
Fix Functions::Monomial::gradient
Fix Functions::Monomial::gradient
Fix Functions::Monomial::gradient
Fix Functions::Monomial::gradient
Fix Functions::Monomial::gradient
Wolfgang Bangerth [Mon, 29 Jun 2015 21:24:14 +0000 (16:24 -0500)]
Handle the case of lucky breakdown better, without testing whether 1/a
is actually finite.
In the beginning the Universe was created. This has made a lot of
people very angry and has been widely regarded as a bad move.
Douglas Adams
Typeset in
Trocchi and Trocchi Bold Sans Serif.